Online prediction of pressing iron ore concentrates in an industrial HPGR. Part 2: Digital assistant

Abstract

Besides the recent advances in high-pressure grinding rolls (HPGR) technology, the good track record demonstrates the potential growth in its applications in the future. In this regard, challenges are emerging driven by technology improvements and the increasing demand for a proper understanding of the operation and optimization of the grinding process in different scenarios. To fill this gap, new dynamic modeling and simulation approaches have been evolving to capture information of the process in real-time using online tools. However, proper application of these models as online digital assistants is still missing. The present work applies the online modeling approach proposed by the authors in the first part of this series as a digital assistant. The digital assistant receives real-time information of an industrial-scale HPGR pressing of iron ore concentrates and uses the Modified Torres and Casali model to predict the main HPGR performance variables. At first, a quantitative description of the HPGR hydro-pneumatic pressurizing system is calibrated and validated describing the relationship between hydraulic pressure and gap in the industrial machine. Feasibility of applying the online model as a digital assistant was demonstrated from two simulation case studies by providing setpoints of hydraulic pressure based on real-time changes in HPGR feed BSA, with the aim of reducing the variabilities of the HPGR product BSA and absorbing a coarser feed BSA.